Bayesian modeling approach for detecting local determinants of the spatial distribution of immigrant communities in large cities: The case of Barcelona

Abstract

This paper presents a novel approach to identify local determinants of the spatial distribution of international immigrants in big cities. The methodology is based on Bayesian spatial analysis, and it is applied to assess the association of a wide range of variables with the location of the main migrant groups in the Barcelona County. The proposed Bayesian modeling strategy allows addressing spatial dependency and feature selection. Overall, the analysis suggests that factors such as distance from the city center, the proportion of rental housing, and access to public transportation account for much of the variation in location choice. However, these factors exhibit varying effects across different immigrant groups, underscoring the heterogeneous nature of settlement patterns. The findings offer valuable insights for urban policymakers seeking to address the challenges posed by immigration in big cities and to design effective integration strategies.